Earlier this year, I had an "if only we could..." thought that was bugging me. We often ask prospective and current clients to challenge us with a sentence that starts with: "if only we could [x]" in the hope that we can surprise them with an answer or a process that turns that aspiration into something along the lines of "I'm so glad we can [x]".
These problems are often around data discovery or enrichment, or the delivery of a hyper-personalised message, and often we can help get closer to the answer than the client expects. I had my own quest, though, and mine was this:
“If only we could find all the UK users of SharpSpring”
We're partners with SharpSpring and we think it's a great marketing automation platform. When new clients come to us without a marketing automation platform, SharpSpring is the one we recommend, although we'll happily work with others if a client already has one in place. One area that we've had success with lately is when a company is already using SharpSpring, but not to its full potential. At that point we're perfectly placed to help that company level up.
With this knowledge, I started to think about how we could find out which UK websites were using SharpSpring, with the end goal being a clear list of potential prospects that we had a likely match with. It didn't take long for that thought to evolve: it would be massively interesting to find out what the overall state of marketing automation usage was among UK websites. How many were using a marketing automation platform in total? How many were using Hubspot, or Marketo, or any of the other numerous solutions?
How we helped PrimeStox treble registrations while reducing spend in just two months.
With my new self-imposed challenge, I started to gather what I'd need to get towards the desired outputs:
With my starting data and methods in place, it was time to write some code. I used Python for this task. I won't go into massive detail on the process, but the code divided the list of websites into chunks and then used the multiprocessing module to check each one for each of the platforms.
The code only checked the homepage of each website – I made the assumption that this page would definitely include the code of any platforms the site was using, and that allowed us to limit both the load on any websites we were visiting as well as keep the number of requests our server was making to a minimum. As the results came in from the code, they were output directly to a .CSV file. No fancy database work for this, although maybe I'd rethink that decision if I was to do this again!
How we developed a unique targeting approach and improved lead volumes from direct marketing by more than 80%.
The process for each website was simple:
I set the process running, went for a cup of coffee, came back to realise it was going to take about two days to run and got on with some actual work. Now that the results are back, let's look at some of the statistics...
We ended up having a list of 678,904 UK websites to start the process with. Of those, 115,634 gave us an error of some sort when we tried to grab the source code of the homepage. That leaves us with an initial source dataset of 563,270 UK business websites to work with. I'm sure that's not an exhaustive list, but I'm happy enough with the size of it to say it'll give us a decent representation of the market as a whole.
OK, so I'm putting the most surprising stat very early on in these results. I expected somewhere around 10% of UK businesses to be using some form of marketing automation platform, but now that we have these results, maybe I am too close to the market to be making those assumptions? As it turns out, just 2.7% of the websites we were able to check showed signs of using one of the platforms we were checking for. For those of us who are in this market, that means over 97% of the businesses that we approach about marketing automation are unlikely to know what we're talking about, let alone have a solution in place already.
This one will depend heavily on your views. Personally, I'd say that Mailchimp can be used for marketing automation activity, but it's extremely unlikely that all users of Mailchimp are using it for more than a regular newsletter. I'd give the top spot here to HubSpot for that reason, but here are the results as-is:
Platform | # Websites | % of MA Users | % of UK websites |
---|---|---|---|
Mailchimp | 9480 | 57.29% | 1.68% |
HubSpot | 3053 | 18.45% | 0.54% |
Pardot | 948 | 5.73% | 0.17% |
Salesforce | 808 | 4.88% | 0.14% |
Zoho | 758 | 4.58% | 0.13% |
Infusionsoft | 450 | 2.72% | 0.08% |
SharpSpring | 398 | 2.41% | 0.07% |
Marketo | 297 | 1.79% | 0.05% |
Act-On | 139 | 0.84% | 0.02% |
Force24 | 112 | 0.68% | 0.02% |
Constant Contact | 100 | 0.60% | 0.02% |
SugarCRM/Salesfusion | 5 | 0.03% | 0.00% |
An interesting way of looking at this data, as far as I'm concerned, is to think about the chance of your message hitting home if you're a service provider aiming at HubSpot users. If you grab a list of UK businesses from somewhere and send them all an email, you've got around a 1 in 185 chance of that email being received by a business who uses HubSpot – and they're the largest of the platforms! Once you add in the added variables of email deliverability, open rates and the like, the "pray" part of "spray and pray" becomes a little more apparent.
The anomaly in these results is the SugarCRM/Salesfusion and Constant Contact numbers. I'd expect both of those to be higher, and so I expect there was an error somewhere in the process.
How we developed a unique targeting approach and improved lead volumes from direct marketing by more than 80%.
As a side note, I'd love to hear from anyone at any of these platforms to let me know whether we're in the right ballpark for the number of UK users you have.
In what is unlikely to be the biggest shock of the decade when it comes to marketing research, the business classification with the highest number of marketing automation users was... Marketing and Advertising Consultants. It should be no surprise that those of us who are most focussed on getting the most from our marketing activity are also the biggest users of this tech. But what about the rest? Here are the top 20 classifications:
Marketing and Advertising Consultants | 346 |
---|---|
Web Design and Development | 331 |
Computer Software Development | 301 |
Training Services | 231 |
Computer Services | 227 |
Recruitment Consultants | 196 |
Telecommunication Services | 153 |
Schools and Colleges | 149 |
Computer Software Sales | 139 |
Educational Services | 128 |
Advertising and Graphic Designers | 123 |
Accountants | 119 |
Internet Services | 115 |
Travel Agents & Services | 111 |
Charitable and Voluntary Organisations | 107 |
Computer Consultancy | 103 |
Newspapers and Magazines | 98 |
Laboratories | 97 |
Financial Advisers | 90 |
Publishers | 87 |
Very interesting, although also not massively surprising, to see that the massively sales-led and numbers-driven businesses in the recruitment and telecoms worlds are high up the list. More surprising to me is the relative number of educational sites that are using a marketing automation platform. Just below this list were some that I would have expected to be in the top group. Classifications like Hotels, Beauty Salons and Estate Agents feel like prime users of automation software and processes, but they don't quite break into the top group.
That's 6.5% of the total number of users of a marketing automation platform. The majority (614) of these results are websites using [Mailchimp and something else], which helps to prove the point that Mailchimp isn't really being used as a marketing automation platform on its own. The other big number is [Salesforce and something else], with 368 websites falling into that group. This one can be explained in two ways: firstly, Salesforce is often used as a CRM first and foremost, so there will often be cases where marketing and sales teams are using software from different stables, but secondly and most importantly Pardot is made by the same company as Salesforce and so is often used in tandem. The combination of [Pardot and Salesforce] makes up over half over the [Salesforce and something else] group.
Interestingly, there are five websites in the results that are treating marketing automation platforms like Panini stickers and collecting them. Those five websites each use four of the platforms and I'm sure they all have great reasons for doing so. I'd say it was because they haven't found SharpSpring yet, but that's not true because two of them have!
This piece of research served it's purpose for us internally, we've discovered data in the form of a list of potential prospects that are perfectly in our sweet spot. We can now contact them with a highly personalised communication that speaks to their likely needs. What has also come from this is a massive reminder that we can't let our proximity to marketing automation software lead us to assume that our potential prospects even know what marketing automationis, let alone be ready to make the leap. I expect that this knowledge will help inform some of our upcoming content and marketing decisions, and help us spread the word among UK businesses about just how powerful a properly implemented marketing automation process can be.
If you have any comments on this process, the findings, or if you just want to talk through your own data discovery or marketing automation needs, then feel free to fire me an email or use any of the other contact methods on my team member page here.
Jason rolls together knowledge of programming, automation and data analysis to provide a high level of technical marketing expertise to The Marketing Eye and its clients.
Technical Director / The Marketing Eye
by Darren Coleshill, 4 minute read
by Darren Coleshill, 3 minute read